Submit to Energies Review for Energies Propose a Special Issue

Journal Browser

► Journal Browser

Machine Learning and Optimization with Applications of Power System II

Special Issue Editors
Special Issue Information
Keywords
Related Special Issues
Published Papers

A special issue of Energies (ISSN 1996-1073). This special issue belongs to the section "F: Electrical Engineering".

Deadline for manuscript submissions: closed (30 June 2020) | Viewed by 32847

Share This Special Issue

Special Issue Editor

Prof. Dr. Hongseok Kim

E-Mail Website
Guest Editor

Department of Electronic Engineering, Sogang University, Seoul 04107, Republic of Korea
Interests: power system with optimal power flow; energy storage; machine learning for energy big data and forecasting; energy trading; microgrids
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

This Special Issue is focused on machine learning and optimization techniques that can be applied for power system operation, such as energy data analytics, time series energy forecasting, renewable energy markets, energy storage systems (ESS), microgrids and distribution networks. Modern power systems face new challenges due to the high penetration of renewable generation, and thus prediction and control are essential for grid reliability. Thanks to massively deployed energy IoT sensors and energy big data, machine learning including deep learning is being actively applied to predict renewable generation and electric loads. The accurate forecasting of PV and wind power is also of prime importance for strategic bidding in renewable energy markets. Deep learning techniques including recurrent neural networks (RNN), long short-term memory (LSTM), and convolution neural networks (CNN) are expected to improve the prediction accuracy of time series energy data.

Nevertheless, forecasting errors are unavoidable, and mitigating the variability of the grid requires other techniques. Indeed, ESS plays a key role in controlling the grid under volatile generation and loads, and is widely deployed for peak cut frequency regulation, bidding in renewable energy markets, demand response, etc. Multiple small scale ESS units can be also aggregated and collectively controlled as one virtual unit. Finally, it is desirable to optimally operate distribution networks and/or microgrids with the aforementioned distributed energy resources; optimal power flow possibly combined with peer-to-peer energy trading is also of great interest.

In this Special Issue, new theoretical and/or practical research results using machine learning and optimization techniques with the application of power systems are solicited. Pilot programs and field tests considering regional requirements are also welcome. The preferred topics include, but are not limited to:

Energy data analytics and forecasting

Deep learning (RNN, LSTM, CNN, etc.) for load and renewable generation prediction

Deep reinforcement learning for stochastic control

ESS operation considering uncertainty, frequency regulation, demand response, and/or battery degradation

Demand response

Energy bidding and game theory in renewable energy markets

Pilot programs and field tests

Microgrid optimization and simulator development

Optimal power flow in distribution networks

Virtual power plants

Dr. Hongseok Kim
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Energies is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

Energy data analytics and forecasting
Deep learning (RNN, LSTM, CNN, etc.) for load and renewable generation prediction
Deep reinforcement learning for stochastic control
ESS operation considering uncertainty, frequency regulation, demand response, and/or battery degradation
Demand response
Energy bidding and game theory in renewable energy markets
Pilot programs and field tests
Microgrid optimization and simulator development
Optimal power flow in distribution networks
Virtual power plants

Related Special Issues

Machine Learning and Optimization with Applications of Power System III in Energies (8 articles)
Machine Learning and Optimization with Applications of Power System in Energies (10 articles)

Published Papers (11 papers)

Download All Papers

Research

17 pages, 6115 KiB

Open AccessArticle

Forecasting Day-Ahead Hourly Photovoltaic Power Generation Using Convolutional Self-Attention Based Long Short-Term Memory

by Dukhwan Yu, Wonik Choi, Myoungsoo Kim and Ling Liu

Energies 2020, 13(15), 4017; https://0-doi-org.brum.beds.ac.uk/10.3390/en13154017 - 04 Aug 2020

Cited by 23 | Viewed by 3447

Abstract

The problem of Photovoltaic (PV) power generation forecasting is becoming crucial as the penetration level of Distributed Energy Resources (DERs) increases in microgrids and Virtual Power Plants (VPPs). In order to improve the stability of power systems, a fair amount of research has been proposed for increasing prediction performance in practical environments through statistical, machine learning, deep learning, and hybrid approaches. Despite these efforts, the problem of forecasting PV power generation remains to be challenging in power system operations since existing methods show limited accuracy and thus are not sufficiently practical enough to be widely deployed. Many existing methods using long historical data suffer from the long-term dependency problem and are not able to produce high prediction accuracy due to their failure to fully utilize all features of long sequence inputs. To address this problem, we propose a deep learning-based PV power generation forecasting model called Convolutional Self-Attention based Long Short-Term Memory (LSTM). By using the convolutional self-attention mechanism, we can significantly improve prediction accuracy by capturing the local context of the data and generating keys and queries that fit the local context. To validate the applicability of the proposed model, we conduct extensive experiments on both PV power generation forecasting using a real world dataset and power consumption forecasting. The experimental results of power generation forecasting using the real world datasets show that the MAPEs of the proposed model are much lower, in fact by 7.7%, 6%, 3.9% compared to the Deep Neural Network (DNN), LSTM and LSTM with the canonical self-attention, respectively. As for power consumption forecasting, the proposed model exhibits 32%, 17% and 44% lower Mean Absolute Percentage Error (MAPE) than the DNN, LSTM and LSTM with the canonical self-attention, respectively. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Graphical abstract

15 pages, 1927 KiB

Open AccessArticle

Direct Normal Irradiance Forecasting Using Multivariate Gated Recurrent Units

by Majid Hosseini, Satya Katragadda, Jessica Wojtkiewicz, Raju Gottumukkala, Anthony Maida and Terrence Lynn Chambers

Energies 2020, 13(15), 3914; https://0-doi-org.brum.beds.ac.uk/10.3390/en13153914 - 31 Jul 2020

Cited by 16 | Viewed by 2374

Abstract

Power grid operators rely on solar irradiance forecasts to manage uncertainty and variability associated with solar power. Meteorological factors such as cloud cover, wind direction, and wind speed affect irradiance and are associated with a high degree of variability and uncertainty. Statistical models fail to accurately capture the dependence between these factors and irradiance. In this paper, we introduce the idea of applying multivariate Gated Recurrent Units (GRU) to forecast Direct Normal Irradiance (DNI) hourly. The proposed GRU-based forecasting method is evaluated against traditional Long Short-Term Memory (LSTM) using historical irradiance data (i.e., weather variables that include cloud cover, wind direction, and wind speed) to forecast irradiance forecasting over intra-hour and inter-hour intervals. Our evaluation on one of the sites from Measurement and Instrumentation Data Center indicate that both GRU and LSTM improved DNI forecasting performance when evaluated under different conditions. Moreover, including wind direction and wind speed can have substantial improvement in the accuracy of DNI forecasts. Besides, the forecasting model can accurately forecast irradiance values over multiple forecasting horizons. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Figure 1

17 pages, 4877 KiB

Open AccessArticle

Frequency Performance Distribution Index for Short-Term System Frequency Reliability Forecast Considering Renewable Energy Integration

by Kofi Afrifa Agyeman, Ryota Umezawa and Sekyung Han

Energies 2020, 13(11), 2945; https://0-doi-org.brum.beds.ac.uk/10.3390/en13112945 - 08 Jun 2020

Cited by 1 | Viewed by 1764

Abstract

Risk in a power system’s ability to survive imminent disturbances without recourse to low operational cost and non-interruptive energy delivery remains the responsibility of every grid operator. Intermittencies in renewable energy and dynamic load variations influence the quality of power supply. The sudden changes affect the system frequency, compromising the reliability of the system grid; generation response to frequency regulation is momentous in such an incident. Slower response or smaller reserve capacity may cause a power shortage. This paper proposes a novel predictive scheme for a short-term operational reliability evaluation for system operations planning. The proposed method evaluates the operational reliability of system frequency whiles considering high renewable power penetration and energy storage system incorporation. Required energy generations, and other grid parameters, are modelled as stochastic inputs to the framework. We formulate a reliability index as a frequency distribution considering system frequency control dynamics and processes. The IEEE Reliability Test System (RTS) is used to prove the efficacy of the proposed model. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Graphical abstract

20 pages, 3069 KiB

Open AccessArticle

Short-Term Photovoltaic Power Forecasting Using a Convolutional Neural Network–Salp Swarm Algorithm

by Happy Aprillia, Hong-Tzer Yang and Chao-Ming Huang

Energies 2020, 13(8), 1879; https://0-doi-org.brum.beds.ac.uk/10.3390/en13081879 - 12 Apr 2020

Cited by 48 | Viewed by 3565

Abstract

The high utilization of renewable energy to manage climate change and provide green energy requires short-term photovoltaic (PV) power forecasting. In this paper, a novel forecasting strategy that combines a convolutional neural network (CNN) and a salp swarm algorithm (SSA) is proposed to forecast PV power output. First, the historical PV power data and associated weather information are classified into five weather types, such as rainy, heavy cloudy, cloudy, light cloudy and sunny. The CNN classification is then used to determine the prediction for the next day’s weather type. Five models of CNN regression are established to accommodate the prediction for different weather types. Each CNN regression is optimized using a salp swarm algorithm (SSA) to tune the best parameter. To evaluate the performance of the proposed method, comparisons were made to the SSA based support vector machine (SVM-SSA) and long short-term memory neural network (LSTM-SSA) methods. The proposed method was tested on a PV power generation system with a 500 kWp capacity located in south Taiwan. The results showed that the proposed CNN-SSA could accommodate the actual generation pattern better than the SVM-SSA and LSTM-SSA methods. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Graphical abstract

12 pages, 3081 KiB

Open AccessArticle

Integration of Smart Grid Resources into Generation and Transmission Planning Using an Interval-Stochastic Model

by Guk-Hyun Moon, Rakkyung Ko and Sung-Kwan Joo

Energies 2020, 13(7), 1843; https://0-doi-org.brum.beds.ac.uk/10.3390/en13071843 - 10 Apr 2020

Cited by 3 | Viewed by 2166

Abstract

In the power industry, the deployment of smart grid resources in power systems has become an issue of major interest. The deployment of smart grid resources represents an additional uncertainty in the integrated generation and transmission planning that raises uncertainties in investment-related decision making. This paper presents a new power system planning method for the integration of electric vehicles (EVs) and wind power generators into power systems. An interval-stochastic programming method is used to account for the heterogeneous uncertainties attributable to natural variability and lack of knowledge. The numerical results compare the multiple integration scenarios and verifies the effectiveness of the proposed method in terms of cost distribution and regret cost. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Figure 1

17 pages, 7445 KiB

Open AccessArticle

Energy Storage System Event-Driven Frequency Control Using Neural Networks to Comply with Frequency Grid Code

by Soseul Jeong, Junghun Lee, Minhan Yoon and Gilsoo Jang

Energies 2020, 13(7), 1657; https://0-doi-org.brum.beds.ac.uk/10.3390/en13071657 - 02 Apr 2020

Cited by 4 | Viewed by 2209

Abstract

As the penetration of renewable energy sources (RESs) increases, the rate of conventional generators and the power system inertia are reduced accordingly, resulting in frequency-stability concerns. As one of the solutions, the battery-type energy storage system (ESS), which can rapidly charge and discharge energy, is utilized for frequency regulation. Typically, it is based on response-driven frequency control (RDFC), which adjusts its output according to the measured frequency. In contrast, event-driven frequency control (EDFC) involves a determined frequency support scheme corresponding to a particular event. EDFC has the advantage that control action is promptly performed compared to RDFC. This study proposes an ESS EDFC strategy that involves estimating the required operating point of the ESS according to a specific disturbance through neural-network training. When a disturbance occurs, the neural networks can estimate the proper magnitude and duration of the ESS output to comply with the frequency grid code. A simulation to validate the proposed control method was performed for an IEEE 39 bus system. The simulation results indicate that a neural-network estimation offers sufficient accuracy for practical use, and frequency response can be adjusted as intended by the system operator. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Figure 1

16 pages, 3135 KiB

Open AccessArticle

An Ensemble Learner-Based Bagging Model Using Past Output Data for Photovoltaic Forecasting

by Sunghyeon Choi and Jin Hur

Energies 2020, 13(6), 1438; https://0-doi-org.brum.beds.ac.uk/10.3390/en13061438 - 19 Mar 2020

Cited by 29 | Viewed by 3418

Abstract

As the world is aware, the trend of generating energy sources has been changing from conventional fossil fuels to sustainable energy. In order to reduce greenhouse gas emissions, the ratio of renewable energy sources should be increased, and solar and wind power, typically, are driving this energy change. However, renewable energy sources highly depend on weather conditions and have intermittent generation characteristics, thus embedding uncertainty and variability. As a result, it can cause variability and uncertainty in the power system, and accurate prediction of renewable energy output is essential to address this. To solve this issue, much research has studied prediction models, and machine learning is one of the typical methods. In this paper, we used a bagging model to predict solar energy output. Bagging generally uses a decision tree as a base learner. However, to improve forecasting accuracy, we proposed a bagging model using an ensemble model as a base learner and adding past output data as new features. We set base learners as ensemble models, such as random forest, XGBoost, and LightGBMs. Also, we used past output data as new features. Results showed that the ensemble learner-based bagging model using past data features performed more accurately than the bagging model using a single model learner with default features. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Figure 1

10 pages, 1159 KiB

Open AccessArticle

Stochastic Mixed-Integer Programming (SMIP)-Based Distributed Energy Resource Allocation Method for Virtual Power Plants

by Rakkyung Ko and Sung-Kwan Joo

Energies 2020, 13(1), 67; https://0-doi-org.brum.beds.ac.uk/10.3390/en13010067 - 21 Dec 2019

Cited by 6 | Viewed by 2105

Abstract

Virtual power plants (VPPs) have been widely researched to handle the unpredictability and variable nature of renewable energy sources. The distributed energy resources are aggregated to form into a virtual power plant and operate as a single generator from the perspective of a system operator. Power system operators often utilize the incentives to operate virtual power plants in desired ways. To maximize the revenue of virtual power plant operators, including its incentives, an optimal portfolio needs to be identified, because each renewable energy source has a different generation pattern. This study proposes a stochastic mixed-integer programming based distributed energy resource allocation method. The proposed method attempts to maximize the revenue of VPP operators considering market incentives. Furthermore, the uncertainty in the generation pattern of renewable energy sources is considered by the stochastic approach. Numerical results show the effectiveness of the proposed method. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Figure 1

17 pages, 5120 KiB

Open AccessArticle

Fault Current Constraint Transmission Expansion Planning Based on the Inverse Matrix Modification Lemma and a Valid Inequality

by Sungwoo Lee, Hyoungtae Kim, Tae Hyun Kim, Hansol Shin and Wook Kim

Energies 2019, 12(24), 4769; https://0-doi-org.brum.beds.ac.uk/10.3390/en12244769 - 13 Dec 2019

Cited by 3 | Viewed by 2294

Abstract

In the transmission expansion planning (TEP) problem, it is challenging to consider a fault current level constraint due to the time-consuming update process of the bus impedance matrix, which is required to calculate the fault currents during the search for the optimal solution. In the existing studies, either a nonlinear update equation or its linearized version is used to calculate the updated bus impedance matrix. In the former case, there is a problem in that the mathematical formulation is derived in the form of mixed-integer nonlinear programming. In the latter case, there is a problem in that an error due to the linearization may exist and the change of fault currents in other buses that are not connected to the new transmission lines cannot be detected. In this paper, we use a method to obtain the exact updated bus impedance matrix directly from the inversion of the bus admittance matrix. We propose a novel method based on the inverse matrix modification lemma (IMML) and a valid inequality is proposed to find a better solution to the TEP problem with fault current level constraint. The proposed method is applied to the IEEE two-area reliability test system with 96 buses to verify the performance and effectiveness of the proposed method and we compare the results with the existing methods. Simulation results show that the existing TEP method based on impedance matrix modification method violates the fault current level constraint in some buses, while the proposed method satisfies the constraint in all buses in a reasonable computation time. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Graphical abstract

14 pages, 846 KiB

Open AccessArticle

Multi-Site Photovoltaic Forecasting Exploiting Space-Time Convolutional Neural Network

by Jaeik Jeong and Hongseok Kim

Energies 2019, 12(23), 4490; https://0-doi-org.brum.beds.ac.uk/10.3390/en12234490 - 25 Nov 2019

Cited by 30 | Viewed by 3660

Abstract

The accurate forecasting of photovoltaic (PV) power generation is critical for smart grids and the renewable energy market. In this paper, we propose a novel short-term PV forecasting technique called the space-time convolutional neural network (STCNN) that exploits the location information of multiple PV sites and historical PV generation data. The proposed structure is simple but effective for multi-site PV forecasting. In doing this, we propose a greedy adjoining algorithm to preprocess PV data into a space-time matrix that captures spatio-temporal correlation, which is learned by a convolutional neural network. Extensive experiments with multi-site PV generation from three typical states in the US (California, New York, and Alabama) show that the proposed STCNN outperforms the conventional methods by up to 33% and achieves fairly accurate PV forecasting, e.g., 4.6–5.3% of the mean absolute percentage error for a 6 h forecasting horizon. We also investigate the effect of PV sites aggregation for virtual power plants where errors from some sites can be compensated by other sites. The proposed STCNN shows substantial error reduction by up to 40% when multiple PV sites are aggregated. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures

Figure 1

24 pages, 4660 KiB

Open AccessArticle

Appliance Classification by Power Signal Analysis Based on Multi-Feature Combination Multi-Layer LSTM

by Jin-Gyeom Kim and Bowon Lee

Energies 2019, 12(14), 2804; https://0-doi-org.brum.beds.ac.uk/10.3390/en12142804 - 21 Jul 2019

Cited by 56 | Viewed by 4671

Abstract

The imbalance of power supply and demand is an important problem to solve in power industry and Non Intrusive Load Monitoring (NILM) is one of the representative technologies for power demand management. The most critical factor to the NILM is the performance of the classifier among the last steps of the overall NILM operation, and therefore improving the performance of the NILM classifier is an important issue. This paper proposes a new architecture based on the RNN to overcome the limitations of existing classification algorithms and to improve the performance of the NILM classifier. The proposed model, called Multi-Feature Combination Multi-Layer Long Short-Term Memory (MFC-ML-LSTM), adapts various feature extraction techniques that are commonly used for audio signal processing to power signals. It uses Multi-Feature Combination (MFC) for generating the modified input data for improving the classification performance and adopts Multi-Layer LSTM (ML-LSTM) network as the classification model for further improvements. Experimental results show that the proposed method achieves the accuracy and the F1-score for appliance classification with the ranges of 95–100% and 84–100% that are superior to the existing methods based on the Gated Recurrent Unit (GRU) or a single-layer LSTM. Full article

(This article belongs to the Special Issue Machine Learning and Optimization with Applications of Power System II)

► Show Figures